AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

TINY BUT MIGHTY: A SOFTWARE-HARDWARE CO- DESIGN APPROACH FOR EFFICIENT MULTIMODAL IN- FERENCE ON BATTERY-POWERED SMALL DEVICES

Yilong Li, Yijing Zeng, Shuai Zhang, Hao Zhang, Xinmiao Xiong, Jingyu Liu, Pan Hu, Suman Banerjee

ICLR 2026 · Architectures and efficiency › RNNs and normalization · approach: Distributed LLM inference systems

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
decompose LMM and offload vision encoder to NPU, LLM to GPU with zero-copy TABMNanomind · edge devices, end-to-end energyenergy: 42.3% reductionImproved
cross-accelerator dynamic offloading vs official rkllm on Orange Pi 5 UltraNanomind · Qwen2-VL-2B-Instruct InfoVQAend-to-end latency: 36.2% reductionImproved
zero-copy TABM ring buffer instead of llama.cpp copy-based layer offloading (ablation)Nanomind · embedding transfermemory usage and CPU loadImproved
run vision encoder on NPU instead of GPU or CPU (ablation)Nanomind · SigLip and ArcFace latencyper-image latencyImproved
fused dequant-GEMM OpenCL kernel for LLM decoding (ablation)Nanomind · Qwen2-1.5B-W8A8 text-onlythroughput (tok/s)Improved
offload more layers to GPU in llama.cpp (ablation)llama.cpp · Llama-3-8B (2-bit)memory usage: 2.9GB -> 5.5GBHurt
lower ViT precision in hybrid quantization (ablation)Nanomind · MMBench, MME, InfoVQAaccuracyHurt
kernelized linear attention instead of quadratic attention on GPU (ablation)Nanomind · tested benchmarksaccuracyNo clear effect
switch to on-demand cascade inference in critical battery stateNanomind · 2000 mAh batteryruntime: up to 18.8 hoursImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy